深度学习赋能地图制图的若干思考

doi:10.11947/j.AGCS.2021.20210091

摘要/Abstract

摘要： 地图制图学包含地图制作与地图应用两大任务，分别与人工智能技术有不解之缘。经历了符号主义智能表达的地图制图专家系统、行为主义智能表达的空间优化决策后，地图制图面临与连接主义下的深度学习的结合，以提升地图制图的智能化水平。本文针对“深度学习+地图制图”命题讨论了3个问题。一是从深度学习方法与地图空间问题解答策略思想的一致性，基于梯度下降、局部相关、特征降维和非线性化性质，回答了两者结合的可行性；二是从地图学独特的学科特点和技术环境分析了两者结合面临的挑战，涉及地图数据组织的非规范性、样本建立的专业需求、几何与地理特征的融合，以及地图固有的空间尺度性；三是分别讨论了地图制作与地图应用融入深度学习的切入点和具体方法。

关键词: 地图制图, 人工智能, 深度学习, 图卷积学习模型

Abstract: The cartography discipline includes issues of map making and map applications. Both tasks have deep associations with artificial intelligence. Among different intelligence representation methods, the symbolism intelligence approach used to apply with cartography generating mapping expert system technology, the activism intelligence applied with map analysis resulting in optimization decision technology. Nowadays the combination of cartography and connectionism intelligence deep learning faces challenging problems to improve the intelligence level. This study focuses on the issue “deep learning+cartography” discussing three questions. First from the perspective of the consistent ideas in deep learning and map space settlement argues the combination is possible, because both methods have the similar ideas of gradient descent, local spatial association, dimension reduction and non-linear processing. Secondly, by analyzing the mapping characteristics and technology contexts discusses the challenges from the combination, including the irregular data structure in map organization, sample establishment requiring geo-domain knowledge, the integration of geometric and geographic properties and the spatial scale issues in cartography. Thirdly, from the viewpoints of map making and map application respectively examines the practical methods to combine deep learning and cartography.

Key words: cartography, artificial intelligence, deep learning, graph convolution neural network

中图分类号:

P208

艾廷华. 深度学习赋能地图制图的若干思考[J]. 测绘学报, 2021, 50(9): 1170-1182.

AI Tinghua. Some thoughts on deep learning enabling cartography[J]. Acta Geodaetica et Cartographica Sinica, 2021, 50(9): 1170-1182.

参考文献

[1] 陆漱芬. 制图学是一门工具科学——对地图应用问题的设想[J]. 测绘学报, 1992, 21(4): 307-311. LU Shufen. Cartography can be regarded as an implemental science[J]. Acta Geodaetica et Cartographica Sinica, 1992, 21(4): 307-311.
[2] 高俊. 地图学四面体——数字化时代地图学的诠释[J]. 测绘学报, 2004, 33(1): 6-11. GAO Jun. Cartographic tetrahedron: explanation of cartography in the digital era[J]. Acta Geodaetica et Cartographica Sinica, 2004, 33(1): 6-11.
[3] 郭仁忠, 应申. 论ICT时代的地图学复兴[J]. 测绘学报, 2017, 46(10): 1274-1283. DOI: 10.11947/j.AGCS.2017.20170335. GUO Renzhong, YING Shen. The rejuvenation of cartography in ICT era[J]. Acta Geodaetica et Cartographica Sinica, 2017, 46(10): 1274-1283. DOI: 10.11947/j.AGCS.2017.20170335.
[4] 闾国年, 俞肇元, 袁林旺, 等. 地图学的未来是场景学吗?[J]. 地球信息科学学报, 2018, 20(1): 1-6. LÜ Guonian, YU Zhaoyuan, YUAN Linwang, et al. Is the future of cartography the scenario science?[J]. Journal of Geo-Information Science, 2018, 20(1): 1-6.
[5] 余卓渊, 闾国年, 张夕宁, 等. 全息高精度导航地图: 概念及理论模型[J]. 地球信息科学学报, 2020, 22(4): 760-771. YU Zhuoyuan, LÜ Guonian, ZHANG Xining, et al. Pan-information-based high precision navigation map: concept and theoretical model[J]. Journal of Geo-Information Science, 2020, 22(4): 760-771.
[6] WEIBEL R, KELLER S, REICHENBACHER T. Overcoming the knowledge acquisition bottleneck in map generalization: the role of interactive systems and computational intelligence[C]//Proceedings of 1995 International Conference on Spatial Information Theory. Semmering, Austria: Springer, 1995: 139-156.
[7] 艾廷华. 适宜空间认知结果表达的地图形式[J]. 遥感学报, 2008, 12(2): 347-354. AI Tinghua. Maps adaptable to represent spatial cognition[J]. Journal of Remote Sensing, 2008, 12(2): 347-354.
[8] 周志华. 机器学习[M]. 北京: 清华大学出版社, 2016. ZHOU Zhihua. Machine learning[M]. Beijing: Tsinghua University Press, 2016.
[9] 孙群. 专家系统以及它在地图制图领域中的应用[J]. 测绘学院学报, 1992(1): 67-73. SUN Qun. Expert system and its application in cartography[J]. Journal of Institute of Surveying and Mapping, 1992(1): 67-73.
[10] 华一新. 用专家系统技术确定地图要素的地图符号类型[J]. 测绘学院学报, 1991(3): 43-47, 55. HUA Yixin. Determine the map symbol type of map element with expert system technology[J]. Journal of Institute of Surveying and Mapping, 1991(3): 43-47, 55.
[11] 张文星, 苏波, 李华, 等. 集成式专家系统工具GEST[J]. 武汉测绘科技大学学报, 1992, 17(3): 1-8. ZHANG Wenxing, SU Bo, LI Hua, et al. An integrated expert system tool-GEST[J]. Journal of Wuhan Technical University of Surveying and Mapping, 1992, 17(3): 1-8.
[12] SESTER M. Knowledge acquisition for the automatic interpretation of spatial data[J]. International Journal of Geographical Information Science, 2000, 14(1): 1-24.
[13] SESTER M. Optimization approaches for generalization and data abstraction[J]. International Journal of Geographical Information Science, 2005, 19(8-9): 871-897.
[14] 钱海忠,武芳,王家耀.自动制图综合及其过程控制的智能化研究[M].北京:测绘出版社,2012. QIAN Haizhong, WU Fang, WANG Jiayao. Study of automated cartographic generalization and intelligentized generalization process control[M]. Beijing: Surveying and Mapping Press, 2012.
[15] 高松. 地理空间人工智能的近期研究总结与思考[J]. 武汉大学学报(信息科学版), 2020, 45(12): 1865-1874. GAO Song. A review of recent researches and reflections on geospatial artificial intelligence[J]. Geomatics and Information Science of Wuhan University, 2020, 45(12): 1865-1874.
[16] TOUYA G, ZHANG X, LOKHAT I. Is deep learning the new agent for map generalization?[J]. International Journal of Cartography, 2019, 5(2-3): 142-157.
[17] LEI Yingzhe, AI Tinghua, ZHANG Xiang, et al. A parallel annotation placement method for dense point of interest labels using hexagonal grid[J]. Cartography and Geographic Information Science, 2021, 48(2): 95-104.
[18] LECUN Y, BENGIO Y, HINTON G. Deep learning[J]. Nature, 2015, 521(7553): 436-444.
[19] LECUN Y, BOTTOU L, BENGIO Y, et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 1998, 86(11): 2278-2324.
[20] REICHSTEIN M, CAMPS-VALLS G, STEVENS B, et al. Deep learning and process understanding for data-driven Earth system science[J]. Nature, 2019, 566(7743): 195-204.
[21] ZHU Di, LIU Yu. Modelling spatial patterns using graph convolutional networks (short paper)[C]//Proceedings of the 10th International Conference on Geographic Information Science. Dagstuhl, Germany: Schloss Dagstuhl-Leibniz-Zentrum fuer Informatik, 2018, 73:1-7.
[22] JENNY B, HEITZLER M, SINGH D, et al. Cartographic relief shading with neural networks[J]. IEEE Transactions on Visualization and Computer Graphics, 2021, 27(2): 1225-1235.
[23] 刘经南, 詹骄, 郭迟, 等. 智能高精地图数据逻辑结构与关键技术[J]. 测绘学报, 2019, 48(8): 939-953. DOI: 10.11947/j.AGCS.2019.20190125. LIU Jingnan, ZHAN Jiao, GUO Chi, et al. Data logic structure and key technologies on intelligent high-precision map[J]. Acta Geodaetica et Cartographica Sinica, 2019, 48(8): 939-953. DOI: 10.11947/j.AGCS.2019.20190125.
[24] MCCULLOCH W S, PITTS W. A logical calculus of the ideas immanent in nervous activity[J]. The Bulletin of Mathematical Biophysics, 1943, 5(4): 115-133.
[25] RUMELHART D E, HINTON G E, WILLIAMS R J. Learning representations by back-propagating errors[J]. Nature, 1986, 323(6088): 533-536.
[26] O’CALLAGHAN J F, MARK D M. The extraction of drainage networks from digital elevation data[J]. Computer Vision, Graphics, and Image Processing, 1984, 28(3): 323-344.
[27] NIEPERT M, AHMED M, KUTZKOV K. Learning convolutional neural networks for graphs[C]//Proceedings of the 33rd International Conference on Machine Learning. New York: Curran Associates, Inc., 2016: 2014-2023.
[28] KIPF T N, WELLING M. Semi-supervised classification with graph convolutional networks[C]//Proceedings of the 5th International Conference on Learning Representations. Toulon, France: ICLR, 2017.
[29] TOBLER W R. A computer movie simulating urban growth in the Detroit region[J]. Economic Geography, 1970, 46(S1): 234-240.
[30] ANSELIN L. Local indicators of spatial association: LISA[J]. Geographical Analysis, 1995, 27(2): 93-115.
[31] 艾廷华. 大数据驱动下的地图学发展[J]. 测绘地理信息, 2016, 41(2): 1-7. AI Tinghua. Development of cartography driven by big data[J]. Journal of Geomatics, 2016, 41(2): 1-7.
[32] REN Shaoqing, HE Kaiming, GIRSHICK R, et al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6): 1137-1149.
[33] HE Kaiming, GKIOXARI G, DOLLÁR P, et al. Mask R-CNN[C]//Proceedings of 2017 IEEE International Conference on Computer Vision. Venice, Italy: IEEE, 2017: 2980-2988.
[34] YU Bing, YIN Haoteng, ZHU Zhanxing. Spatio-temporal graph convolutional neural network: a deep learning framework for traffic forecasting[C]//Proceedings of the 27th International Joint Conference on Artificial Intelligence Main Track. Stockholm: IJCAI, 2018: 3634-3640.
[35] XING Hanfa, MENG Yuan. Integrating landscape metrics and socioeconomic features for urban functional region classification[J]. Computers, Environment and Urban Systems, 2018, 72: 134-145.
[36] CAO Rui, TU Wei, YANG Cuixin, et al. Deep learning-based remote and social sensing data fusion for urban region function recognition[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2020, 163: 82-97.
[37] 李志林, 王继成, 谭诗腾, 等. 地理信息科学中尺度问题的30年研究现状[J]. 武汉大学学报(信息科学版), 2018, 43(12): 2233-2242. LI Zhilin, WANG Jicheng, TAN Shiteng, et al. Scale in geo-information science: an overview of thirty-year development[J]. Geomatics and Information Science of Wuhan University, 2018, 43(12): 2233-2242.
[38] PLAZANET C, BIGOLIN N M, RUAS A. Experiments with learning techniques for spatial model enrichment and line generalization[J]. GeoInformatica, 1998, 2(4): 315-333.
[39] RUAS A, DUCHÊNE C. A prototype generalisation system based on the multi-agent system paradigm[M]//MACKANESS W A, RUAS A, SARJAKOSKI L T. Generalisation of Geographic Information. Amsterdam: Elsevier, 2007: 269-284.
[40] SESTER M, FENG Yu, THIEMANN F. Building generalization using deep learning[C]//Proceedings of the International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Delft: [s.n.], 2018: 565-572.
[41] YAN Xiongfeng, AI Tinghua, YANG Min, et al. Graph convolutional autoencoder model for the shape coding and cognition of buildings in maps[J]. International Journal of Geographical Information Science, 2021, 35(3): 490-512.
[42] YAN Xiongfeng, AI Tinghua, YANG Min, et al. A graph convolutional neural network for classification of building patterns using spatial vector data[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2019, 150: 259-273.
[43] LEE J, JANG H, YANG J, et al. 2017. Machine learning classification of buildings for map generalization[J]. ISPRS International Journal of Geo-Information, 2017, 6 (10), 309-324.
[44] GATYS L A, ECKER A S, BETHGE M. Image style transfer using convolutional neural networks[C]//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, NV: IEEE, 2016: 2414-2423.
[45] GOODFELLOW I J, POUGET-ABADIE J, MIRZA M, et al. Generative adversarial networks[C]//Proceedings of the 27th International Conference on Neural Information Processing Systems: vol. 2. Cambridge MA: MIT Press, 2014: 2672-2680.
[46] SCHNVRER R, SIEBER R, SCHMID-LANTER J, et al. Detection of pictorial map objects with convolutional neural networks[J]. The Cartographic Journal, 2020.
[47] 任加新, 刘万增, 李志林, 等. 利用卷积神经网络进行“问题地图”智能检测[J]. 武汉大学学报(信息科学版), 2021, 46(4): 570-577. REN Jiaxin, LIU Wanzeng, LI Zhilin, et al. Intelligent detection of “Problematic Map” using convolutional neural network[J]. Geomatics and Information Science of Wuhan University, 2021, 46(4): 570-577.
[48] 王米琪, 艾廷华, 晏雄锋, 等. 图卷积网络模型识别道路正交网格模式[J]. 武汉大学学报(信息科学版), 2020, 45(12): 1960-1969. WANG Miqi, AI Tinghua, YAN Xiongfeng, et al. Grid pattern recognition in road networks based on graph convolution network model[J]. Geomatics and Information Science of Wuhan University, 2020, 45(12): 1960-1969.
[49] 何海威, 钱海忠, 谢丽敏, 等. 立交桥识别的CNN卷积神经网络法[J]. 测绘学报, 2018, 47(3): 385-395. DOI: 10.11947/j.AGCS.2018.20170265. HE Haiwei, QIAN Haizhong, XIE Limin, et al. Interchange recognition method based on CNN[J]. Acta Geodaetica et Cartographica Sinica, 2018, 47(3): 385-395. DOI: 10.11947/j.AGCS.2018.20170265.
[50] HU Sheng, GAO Song, WU Liang, et al Urban function classification at road segment level using taxi trajectory data: a graph convolutional neural network approach[J]. Computers, Environment and Urban Systems, 2021, 87: 101619.